About me:

A compassionate, enthusiastic and team player, Cyril is able to prioritize and bring solutions to the technology world. I have 4 years hands on experience with C# and .NET . I strive to learn new things everyday and put it into practice to make the world a better and simple place to live. I have a keen interest in JavaScript, Python, PHP, R, C# and Java. Data Science is another field I am interested in especially Machine Learning and Deep Learning.

Education:

Grand Valley State University (2024 - Current)
Masters Degree in Data Science and Analytics

Strathmore University (2016 - 2019)
Bachelor’s Degree in Informatics and Computer Science

Reflections:

My growth

Since we started learning this course this semester, I feel I am proficient in the following areas of the course work: Regression (linear and logistic regression), single and multiple variable regression, and inference. My introduction to regression gave me an impression of machine learning, but after reading the study materials and watching the “to watch” videos, I understood the basic difference between statistical learning and machine learning. Furthermore, I was able to understand the concepts of generalized linear regression models. Contrary to other coding tools, I found it interesting to use available packages such as tidymodels and GLMnet to build models with less code, thus focusing more on addressing the task at hand than writing code from scratch.

From this course, I was able to establish a clear connection on how concepts such as discriminant analysis (a concept i studied in STA 518) is useful in statistical modeling especially on feature selection. Furthermore, I was able to understand further understand the concept of cross-validation and by using different strategies such as k-fold cross-validation whereby different samples of data were used in evaluating the accuracy of the model in generating predictions. On the other hand, in ensuring that the best model is selected, the tutorials on forward and backward step-wise selection of features were helpful in understanding how important variables that best fit the model can be selected and how to drop unimportant variables to ensure a reliable and accurate model is selected (objective 3).

Cyril Owuor


About me:

A compassionate, enthusiastic and team player, Cyril is able to prioritize and bring solutions to the technology world. I have 4 years hands on experience with C# and .NET . I strive to learn new things everyday and put it into practice to make the world a better and simple place to live. I have a keen interest in JavaScript, Python, PHP, R, C# and Java. Data Science is another field I am interested in especially Machine Learning and Deep Learning.

Education:

Grand Valley State University (2024 - Current)
Masters Degree in Data Science and Analytics

Strathmore University (2016 - 2019)
Bachelor’s Degree in Informatics and Computer Science

Reflections:

My growth

Since we started learning this course this semester, I feel I am proficient in the following areas of the course work: Regression (linear and logistic regression), single and multiple variable regression, and inference. My introduction to regression gave me an impression of machine learning, but after reading the study materials and watching the “to watch” videos, I understood the basic difference between statistical learning and machine learning. Furthermore, I was able to understand the concepts of generalized linear regression models. Contrary to other coding tools, I found it interesting to use available packages such as tidymodels and GLMnet to build models with less code, thus focusing more on addressing the task at hand than writing code from scratch.

From this course, I was able to establish a clear connection on how concepts such as discriminant analysis (a concept i studied in STA 518) is useful in statistical modeling especially on feature selection. Furthermore, I was able to understand further understand the concept of cross-validation and by using different strategies such as k-fold cross-validation whereby different samples of data were used in evaluating the accuracy of the model in generating predictions. On the other hand, in ensuring that the best model is selected, the tutorials on forward and backward step-wise selection of features were helpful in understanding how important variables that best fit the model can be selected and how to drop unimportant variables to ensure a reliable and accurate model is selected (objective 3).